Identification and Discrimination of the Limb Motions using Brain Waves from Motor Imagery

摘要

This research identified and distinguished limb motions by the use of Neural Networks (NN), Support Vector Machines (SVM) and the Electro-encephalogram (EEG). EEG enabled motor imagery of the limbs. By comparing the EEG power at rest with that of the limb's motor imagery, measurements from the electrodes and EEG frequency bands were examined. In this study, EEG with a limited frequency band (from 25Hz to 30Hz) obtained from the electrodes C3, C4, P3 and P4 was used. The features were extracted by the use of Fast Fourier Transform (FFT). The results showed that three out of four subjects were able to identify and distinguish their limb motions at a success rate of over 70%. Furthermore, the rates of identification and discrimination of the limb motions were slightly higher for the Support Vector Machines than for the Neural Networks.